|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: ML_max ado question
Brooks Taggert J <[email protected]>
> So I'm using the hurdle model from Stata Journal Vol 3, No2.
>
> program hurdle_ll
> ...
> ...
> end
>
> ml model lf hurdle_ll [equation here], svy subpop (if [variable]=3D=3D1)
>
> ml max
>
> The first time through it runs fine, the second time it gives me the
> following error:
>
> ML_subv already defined.
>
> I traced it to the ml_max.ado, and it looks like it stems from the svy
> subpop () specification. How come its not being overwritten?
>
> I seem to have found an ugly work around where I just drop it before the
> next run of the model. But clearly there has to be something prettier?
> I imagine Hilbe's hurdle models might be a better idea, but alas I'm
> working in version 8.2.
>
> Any thoughts?
TJ found a bug in -ml-. Fortunately, the fix will be ready in an ado-file
update; unfortunately, ado-file updates are for Stata 9.
Stata 8 users will have to add the following line to ml_clear.ado:
capture drop $ML_subv
and place the modified copy in the 'UPDATES' directory (but not in the 'm'
subdirectory; keep the original asis). Make sure your Stata 8 is fully
up-to-date first. In Stata type
. sysdir
to see where 'UPDATES' is on your machine.
Note that using -ml- non-interactively does not have this problem, so try
coding your call to -ml- with the -max- option, as in
ml model lf hurdle_ll [equation here], ///
svy subpop(if [variable]==1) max
--Jeff
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
